Stronger than carbon nanotubes:
Polyynes and the prospects for carbyne

by Eric Drexler on 2010/09/29


Polyyne rod, 44 carbon atoms + end caps

Carbon nanotubes have a reputation for being strongest possible fibers, but polyyne chains are stronger, as measured by the critical strength/density ratio: Polyyne carbon-carbon bonds are stronger than the bonds in graphene and nanotubes, and the bonds are all are aligned with the axis of the fiber, the optimal geometry for carrying tensile stress. A paper in Nature Chemistry reports the longest polyyne chain synthesized to date, a chain of 44 carbon atoms.

Chemical structure of a polyyne
The longest polyyne to date
(See image at top)

Polyyne structures look like the middle section of the diagram to the right. From a tensile-strength point of view, this standard chemical diagram (which shows alternating single and triple bonds) is somewhat deceptive: What are formally “single bonds” (and therefore might be judged weak) are uncommonly short and strong: < 0.139 nm vs. > 0.142 nm for the bonds in graphene and carbon nanotubes (which have a formal bond order of 1.33).

I’m tempted to cite ~ 0.131 nm for the length of the polyyne bond, but as noted in a recent paper, conventional density-functional calculations are poor at reproducing geometries affected by the Peierls instability, and bond-length alternation in polyynes is a textbook example of it. (Fortunately, this also dampens my urge to run a decent DFT calculation on a polycumulene-like structure that would maintain strong bonding through the interface to an end-group, a cyclopropane. A quick look was interesting, but the interesting parts are linked to the Peierls instability.)

There are reports of a carbon allotrope, “carbyne”, that is (or would?) consist entirely of polyyne chains, but last time I checked (as I do every few years) these reports haven’t described reproducible, well-characterized materials. Carbyne allotropes of carbon are interesting from an advanced-technology perspective, since materials with extreme properties are often useful in engineering, but for now, the instability of polyynes in the laboratory places such stuff outside the scope of conservative exploratory engineering methodologies.

Note the extraordinarily bulky groups at the ends of the newly reported polyyne molecule. They are there for a reason, and the reason is stability.

{ 7 comments… read them below or add one }

Scott Jensen September 30, 2010 at 6:07 pm UTC

So could this be used to build a space elevator from geostationary orbit to surface?


Hi Scott,

If materials based on this could be made, and made in a stable form, then they could outperform a similar material made of carbon nanotubes. Unfortunately, however, I haven’t seen a proposed design for a space elevator that includes a credible failure analysis that addresses what happens when fiber breaks, and so I am not persuaded that the idea can be made to work in the first place.

The fibers in a space elevator would be under enormous stress and would store an enormous amount of elastic energy. Breaking a fiber would convert the elastic energy into kinetic energy (not into heat!). Think of shooting a rubber band, but at the speed of a rifle bullet. What would happen to the adjacent, highly stressed fibers when hit by bullet-speed debris? I’d expect them to break, producing more bullet-speed debris, breaking yet more fibers, and leading to a catastrophic, cascading failure.

When I see a design and analysis that shows how to avoid this problem, I’ll take space elevators somewhat more seriously. Until then, there isn’t a credible design, even at what I would call a conceptual or exploratory level.

— Eric
…………….

Exercise for student readers:

  1. In a typical space elevator cable design, how fast would “bullet speed”, be approximately?
  2. If the kinetic energy were thermalized, how hot would the material be?

(Include input assumptions.)

Jeffrey Soreff October 1, 2010 at 7:41 pm UTC

Neat! Can cyclobutane spacers stabilize poly-cumulene strands without excessively degradation to their stiffness and strength?


Hi Jeff — What sort of spacers do you have in mind?

Jeffrey Soreff October 2, 2010 at 10:35 pm UTC

Hi Eric,
something like:
CH2
/ \
C=C=C C=C=C…
\ /
CH2

Jeffrey Soreff October 2, 2010 at 10:54 pm UTC

Let me try that again… (‘scuse the “.” used as filler – I don’t see how to do
ascii graphics here)
……….C
……….||
……….C
……../….\
….CH2…CH2
……..\…./
……….C
……….||
……….C

The idea being that each cyclobutane spacer makes the cumulene structure
locally energetically favorable (vs. the carbyne structure, with the difference
being roughly the energy to break one bond and create one pair of radicals).
I’m guessing that if the spacers were close enough, the cumulene structure
(with its more-nearly equivalent bonds) would be strongly favored – but at the
cost of introducing the bending stiffness of the sp3 bonds in the cyclobutane
group. I’m guessing that there is some optimum spacing for maximum stiffness.
(To put it another way, the cyclobutane spacers are two of your cyclopropane
terminators merged together, except that I hydrogenated one bond to keep the
overall structure linear).


Hi Jeff!
Thanks, I see what you mean now. Yes, this looks like a good way to divide a carbyne/cumulene chain into manageable segments without sacrificing strength at the junctions; in engineering solid-phase structures, it’s similar to the cyclopropane version and perhaps better for some (or many) hypothetical purposes.

However, a bit of unreliable QC suggests what I’d expect anyway — that the bond-length alternation characteristic of carbyne sets in within a few bond-lengths of the ends, regardless of the terminal bonding.
— Eric

William L. Dye ("willdye") October 3, 2010 at 10:57 am UTC

Are calculations available on how strong a meter-scale cable made of these fibers would be in terms of Pascal’s? I don’t have a subscription to the full article. I’ve read estimates of 50 GPa for carbyne cable, but that was in the mid-90′s.

Eniac October 12, 2010 at 10:00 pm UTC

One way to design a space elevator would be to have the fibers arrayed at a safe distance from each other, in the shape of a wide ribbon. If one breaks, the others pick up the slack. This arrangement also minimizes the meteorite cross section and vulnerability. I am not sure what a safe distance is, but there ought to be one. If a sufficient fraction of the energy is thermalized upon contraction, it could even work in our favor in that the broken fiber might vaporize before it can hit anything.

The bulky endcaps look like they might be useful to make a material, as they would bump into each other and keep the material from creeping, or could be cross-linked. They would also serve as spacers to keep the rods from reacting with each other. Of course, the connection between bump and rod would be the weakest link under tension, which could be a problem.


Hi, Eniac — There may be designs along these lines that would work, but they’d have to be reliable in deflecting the debris from broken fibers away from everything else. Also, to limit the propagation of damage along the length of the structure, it seems that fibers must form segments with ends designed both to transfer load and to break away cleanly on failure.

To claim to be realistic, a design must address this problem in reasonable detail.

(BTW, the tension forces would simply accelerate a broken fiber; there’d be little heating until the debris struck something.)

Roko November 19, 2010 at 12:28 am UTC

Re: space elevators

you can use a hoytether (google) to make a cable out of smaller sub-cables. If you make the individual units of cable small enough (both in diameter and in length) then eventually each one might take such a small fraction of the load that the energy released when it is severed is not serious.

Leave a Comment

Previous post:

Next post: